The neural retina is composed of seven major cell types. The mechanism by which this diversity of retinal cell types develops is unknown. Cells in the developing retina undergo division in the mitotic layer adjacent to the primitive ventricle. Postmitotic cells migrate from the mitotic layer towards the inner surface of the retina where they develop into the various retinal cell types. It has been generally believed that cells migrate uncommitted to a particular phenotype and then at the end of migration, based on various local influences, the cells become committed. However, we have evidence that cells, at least the retinal ganglion cells, begin to differentiate prior to departure from the mitotic layer based on expression of a ganglion cell specific protein, RA4. This suggests that cells become committed to a particular phenotype prior to leaving the mitotic layer. This proposal will develop the hypothesis that retinal cells become committed to a particular phenotype during cell division as a result of interactions with cells that have already differentiated. This hypothesis will be tested in four experiments. The first study will demonstrate that differentiated cells determine the fate of uncommitted cells in vitro. This will involve culturing a mixture of young undifferentiated retinal cells and retinal cells of various ages. The differentiation of the young cells will be followed. The second study will determine if cell division is required for the plasticity of cell phenotype observed in the first experiment. BrdU will be used to label dividing cells in the reaggregate cultures. It will then be possible to determine whether the E4 cells that did or did not differentiate into ganglion cells in the different culture environments underwent division in culture. The third study will determine how soon after mitosis ganglion cells begin to differentiate. This will be accomplished by injecting BrdU into embryos during the peak period of ganglion cell birth to label dividing cells. The time between BrdU injection and the first appearance of BrdU positive mitotic cells will be determined. Next the time between BrdU injection and the first appearance of BrdU positive / RA4 positive cells will be determined. The difference between these two times approximates the shortest interval between mitosis and overt ganglion cell differentiation. This interval will give some indication as to whether the ganglion cell phenotype is induced during or following cell division. The fourth study will clone and characterize the gene encoding for the RA4 protein using a cDNA expression library. This will allow studies on the regulation of the RA4 gene in the future, which may ultimately lead to an insight into the regulation of the ganglion cell phenotype.